Benefit–cost analysis of a low-impact development design

Unpredictable climatic change because of global warming affected droughts and heavy floods (Hu et al. 2017). The dynamic of urban development has altered the natural hydrological cycle and increased the proportion of waterlogged regions. Some extreme situations, such as heavy rains can cause floods around the world, including Indonesia, namely Jakarta (Kardhana et al. 2022) and Semarang (Mudiyono 2022). All these years, excess water such as floods is considered a disaster that dealt with grey infrastructure such as piped conveyances to collect and convey stormwater to wastewater treatment facilities or into surface waters. The purpose is to quickly transport runoff through pipes away from the city to prevent damage to the built environment and avoid insects, disease, and odor caused by stagnant water. This approach may cause overflows and imply a decline in water quality and stream habitats, an increase in stream erosion, and the potential for falling base flows. Therefore, a new system is needed to accommodate the flow into something meaningful. Green infrastructure (GI) (Hanna & Comín 2021), sustainable urban drainage system (SUDS) (Bailey et al. 2019), sponge city (Zha et al. 2021), or low-impact development (LID) is an alternative management that can be applied in conducting water conservation (Bigurra-Alzati et al. 2021). LID was first introduced in North America as a method of engineering design and land planning to control stormwater flow. In recent years, it has gained popularity in urban planning and water resource management due to its ability to replicate pre-development watershed hydrological regimes through infiltration, filtration, storage, evaporation, and containment of runoff near its source (Putri et al. 2023). According to several studies, the use of LID methods can have an impact on water conservation and flood control, including runoff volume (Mai et al. 2018), runoff ratio to rainfall (Hou et al. 2020), and peak flow rate (Baek et al. 2020). Commonly used LID practices are LID water storage (WS) such as rain barrels (RBs) and LID infiltration such as bioretention and vegetative swale (VS) (Jemberie & Melesse 2021). Even though numerous studies have indicated that LID can be used for urban water conservation (Lu & Wang 2021; Zúñiga-Estrada et al. 2022), a better understanding of the maintenance cost-effectiveness throughout the life cycle of a LID is needed to achieve an optimal LID result (Erickson et al. 2018).

The cost-effectiveness of LID performance includes a life cycle cost (LCC) analysis, a technique for determining the most cost-effective choice by adding up all the costs that an object will incur or can be assumed to incur over the course of its service life (Yang et al. 2020). So far, LCC calculations only consider runoff reduction as a benefit of LID (Zeng et al. 2020; Lu et al. 2022). Conversely, as mentioned before the water excess is having other benefits, especially for water conservation and not being wasted. For instance, a RB can be functioned to substitute drinking water used in non-potable water applications such as irrigation and toilet flushing (Oberascher et al. 2021). Bioretention and VSs also can increase the groundwater recharge (Gülbaz & Kazezyılmaz-Alhan 2018) to prevent impacts from groundwater pumping, such as dry wells or sinking lands (Hanak et al. 2019). Besides that, LID is helpful as flood risk protection that can minimalize the socio-economic post-disaster effects by reducing its risk (Sarma & Rajkhowa 2021).

Each LID practice types have different advantages that adapt to area conditions. Beside LID design, other factors that affect the profit of LID are the amount of rain (Kaykhosravi et al. 2018). However, the effects of LID under different rainfall characteristics have not been fully understood, including the economic benefits (Peng et al. 2019). Indonesia faces varying rainfall depth due to the geographics of the Indonesian archipelago laying between two continents and two oceans, crossed by the equator (Priambodo et al. 2019). Different rainfall patterns will have different impacts. Therefore, it is crucial to consider the diverse rainfall depth to obtain the optimal result of LID, especially in Indonesia, which is still considering the utilization of LID (Putri et al. 2023). Thus, this study investigates combining three different types of LID practices with different rainfall depth scenarios to (1) evaluate the hydrology performance of two LID scenarios, namely LID WS and LID Infiltration (I) based on the rainfall depth, and to (2) analyze the LCC based on the rainfall depth to various kinds of its benefits to find the maximum benefits. In this paper, the WS uses RBs, while scenario I uses bioretention and VS. The rainfall–runoff using Storm Water Management Model (SWMM) modeling with four different rainfall depths, which is 20, 25, 30, and 35 mm. This study can be a guidance for decision-makers to implement more targeted and effective sustainable water management strategies across the landscapes with similar rainfall characteristics.

This study used two types of LIDs, namely LID WS in the form of RBs and LID infiltration in the form of VSs and bioretention cells (BCs). The bioretention system itself consists of a column containing surface layer, soil layer, storage layer and drain system with parameters shown in Table 2 as these provide optimum results based on the study by Gülbaz & Kazezyılmaz-Alhan (2017). This study used 78 RBs in 25 sub catchments only used in permanent buildings and main buildings and allowed RBs to be installed. The utilization of LID WS will be analyzed based on the depth of rainfall which is referred to as scenario WS1 (LID WS 20 mm), WS2 (LID WS 25 mm), WS3 (LID WS 30 mm), and WS4 (LID WS 35 mm). Swales are used only on the field in sub-17 and sub-19 and the using of bioretention is 20% of the area in each sub catchment. Like the WS scenario, LID infiltration is also divided into four scenarios based on the rainfall depth, namely I1 (LID infiltration 20 mm), I2 (LID Infiltration 25 mm), I3 (LID infiltration 30 mm), and I4 (LID Infiltration 35 mm). Due to the characteristics of the research region, which is in urban areas and adjacent to the Antirogo river, which is also positioned in the middle of the city as an outflow, SWMM was employed in this study to estimate runoff.

To analyze the benefit–cost in each LID scenario, it is necessary to calculate each benefit and cost component.

The benefits of reducing and utilizing urban rainwater are numerous. According to (Liu et al. 2016), there are seven benefits of using LID were based on local conditions, this study selected six types of benefits with parameters as in Table 3. The calculation method for this benefit is summarized as follows:

The costs incurred in this activity consist of a construction fee with parameters as in Table 4, and operational and maintenance fees of 10% of each benefit.

The decision criteria, if BCR is greater than 1, the project investment was accepted.

The calculation of benefits is carried out by the benefit–cost ratio parameter (Table 3) using Equations (2)(9), which is multiplied by the probability of rain events in each LID scenario. The implementation of both LID scenarios can reduce expenses associated with water resource conservation. The benefits of the B1 aspect are only obtained by using the WS scenario because in scenario I rainfall infiltrated to the ground so that it cannot be used as a substitute for taps such as LID WS, which uses RBs in storing water to be used as a replacement for tap water. According to (Tian et al. 2020), to save expense on tap water and ease the strain on municipal water supplies, the storage pond's rainwater collection system can be used for domestic miscellaneous water use. B2 has the greatest advantage in scenario WS because scenario I does not offer WS, whereas scenario WS allows irrigation at any time. According to (Kim et al. 2022), the benefits of WS on RBs, especially for irrigation, are not limited to that area alone, however, depending on water use habits and storage capacity, RBs can be easily shared with neighbors for a variety of purposes. Therefore, the using of scenario WS can provide benefit not only to the Jember University community but also to the communities surrounding.

On the other hand, aspects B3 to B7 using scenario I provide greater benefits than the WS scenario. This is in line with the ability of scenario I which can reduce runoff and increase the rate infiltration rate better. The B7 aspect is the biggest benefit that can be obtained because scenario I will trigger natural chemical and biological transformations that can naturally reduce pollutants. Meanwhile, in the WS scenario, the ingress of water into the RB causes the amount of runoff to decrease where the less the lymph, the less pollutant content enters the water body. According to (Miller & Cardamone 2021) one of the virtues of using RBs is to help reduce soil runoff caused by reduced runoff so that a slower flow rate causes reduced erosion. The structure of the constituent tissues of LID can also be considered for filtering suspended materials. In general, the use of scenario I provide greater benefits compared to the WS scenario, which can be seen from the net income (Table 5). The greatest benefit occurs in 20 mm of rainfall where its presence is the most frequent.

Investment costs consist of the cost of creating an infrastructure based on the parameters of the LID building used (Tables 2 and 4) and operating and maintenance costs. Construction costs consist of the work on the LID itself, Cf₀, and VAT of 10% (the current tax in Indonesia). Operating and maintenance costs are obtained from 10% of the benefits. So that the investment in scenario I is greater than the WS scenario, where scenario I require construction fees, operations, and maintenance fees that are greater than the WS scenario. This can be explained by the area of work in the scenario I which is also 164% larger than the WS scenario (Table 6). In addition, in scenario I using 2 facilities, namely VSs and BCs with a complex process from soil ranging, structural wall brick, gravel, plant soil management, and plant planting itself. On the other hand, the WS scenario only includes the installation of gutters, pipes, and the installation of the barrel. Based on study results from (Yang et al. 2020) demonstrates how the many structures involved considerably affect the engineering expenses of LID practice. More specifically, RB is the least expensive LID technique since it consists of a straightforward barrel with several pipes and is simple to install and maintain. The largest excavation and material need during the construction phase make PP the most expensive of the three LID techniques.

LID facilities still use conventional labor and concrete building in today's world. High initial investment prices are the result of the absence of industrial and professional development, as well as the adoption of new materials and technologies. However, as suppliers and contractors gained expertise, engineers, architects, and landscape designers continued to refine the design and increase the use of GI strategies to lower construction costs (Bryant 2018). When integrating LID and GI techniques, planners and engineers must overcome numerous challenges. Other hurdles are based on perception and can be removed through education, outreach, and coordination. Some of these barriers are physical and must be overcome with special laws and/or designs. Misperceptions about how LID functions, expenses, upkeep, and waiting for permission clearances are four of the most frequent impediments (Hart et al. 2019).

According to (Abdelhady 2021), a positive present net value suggests an economically viable project, while a negative NPV indicates an economically unsuited enterprise. Present value is a measure of a project's economic viability. According to (Schneider-Marin et al. 2022), the LCC is an economic index that accounts for all lifetime expenses; nevertheless, the LCC is an NPV. These expenses mainly consist of capital, operating, and maintenance costs (Wang et al. 2019).

The largest NPV value occurs in the WS scenario compared to I, especially at a rainfall depth of 20 mm (Table 7), this is influenced by inflows and outflows in the present value. Of the 8 scenarios, scenarios I2, I3, and I4 show negative values or unacceptable projects. The benefit–cost ratio is obtained by dividing the costs incurred by the profits so that the highest benefit–cost occurs in WS1. Based on the payback period, in the scenario I, it takes a much longer time to return capital, especially in scenario I4, compared to scenario WS which only takes 1.5–2 months. The same is the case with BEP where the WS scenario occurs much faster than the scenario I which reaches 11 to almost 12 years. Then it can be determined the magnitude of the cost-effectiveness of the LID design based on the size of the NPV, B/C, payback period, and BEP, where the best scenario is the WS scenario, especially WS1.

High hydrological performance does not equate to high LID facility cost-effectiveness. Although the scenario I in this analysis performed most hydrologically in terms of reducing runoff, it was less cost-effective than utilizing the WS scenario. This is since the region and location of the implementation have a significant impact on the hydrological performance of LID. The available locations are always scarce in built-up urban regions. The study area for the scenario I is significantly larger than the study area's land for scenario WS (Table 7). Similar to this, of the four LIDs examined by (Jiang & McBean 2021), namely permeable pavement (PP), VS, BC, and RB, BC showed the best ability to reduce peak runoff levels and surface runoff volume at many levels, while BC showed the same ability to reduce runoff volume but little peak flow reduction. It turns out that the single-action preference sequence of LID according to comprehensive benefits is: bioretention > RBs > low-elevation greenbelt > green roofs > PP (Li et al. 2017). The hydrological performance per unit of cost, without consideration for the implementation location, is what is meant by ‘cost-effectiveness’. According to the benefits received (Table 6), the scenario I, particularly scenario I1, has the highest net income. Even so, the use of the WS scenario is also no less profitable, where the net income earned ranges from $66,617.49 to $363,749.76. When viewed from the amount of investment spent, the WS scenario requires a much smaller amount of funds than the scenario I, which is 0.05% of the scenario I. Like operating and maintenance costs, the use of the WS scenario requires minimal operational and maintenance costs compared to scenario I, which is 71% of scenario I. In scenario WS1, the return on capital occurs after 1 month and 28 days, while the return on capital in scenario I1 occurs after 5 years 5 months 16 days. Similarly, the break event point in scenario WS1 is achieved much faster than scenario I1, so the cost-effectiveness in scenario WS, especially WS1, is better than scenario I. In scenario I, only I1 can show a positive value, while scenarios I2, I3, and I4 show negative values which means the project is not worth working on. Thus, scenarios I2, I3, and I4 have the highest hydrological performance but the lowest cost-effectiveness. The LID scenario using RB with a 25 mm depth of rainfall is the best scenario with the quickest payback time and least amount of capital but with the greatest advantages. Consequently, in this field, based on the benefits received, the investment made, and the rate of return on capital, the priority is the WS1 > WS2 > WS3 > WS4 > I1 scenario.

However, this study has not explored a combination of WS and scenario I. For example, in a combination of WS and scenario I, both available areas of 74,426.00 m² should be covered by RBs. The combination area maybe 100% WS scenario + 25% scenario I; 100% WS scenario + 50% scenario I or even both scenarios 100%. The RB is very effective to reduce runoff (Yang et al. 2020), along with giving other beneficial opportunities. Based on (Ghodsi et al. 2021), installing gutters that are directly connected to RBs is another way to get LID benefit from combining the RBs with green roofs practice. According to (Mao et al. 2017), the most economical way to fulfill control objectives is using multiple forms of LIDs, including porous pavements, biological retention, and green roofs. Additionally, restoring the natural water cycle is one of the key objectives of LID to conserve water. Rainwater harvesting is anticipated to help improve water supply as well as the restoration of the urban water cycle and a decrease in water-related disasters (Oral et al. 2020). It is advised that LID be implemented on every piece of land that is available (Mai et al. 2018).

The water conservation effectiveness of LID facilities degrades with increasing amounts of rainfall, according to variations in the ratio of runoff volume reduction under various depths of precipitation. LID facilities are less effective in stronger storms, according to similar research (Hu et al. 2019). However, most LID techniques that have been created so far are only useful for modest flood peaks. Additionally, they frequently fail because of the subpar weather patterns at locations and different periods. The optimum LID procedures for the area of interest must be found, technical field efficiency must be raised, and site-specific LID parameter optimization must be optimized (Pour et al. 2020). Here, Cullen and Frey charts (Figure 3) are used for the rain distribution of the Jember region with the Beta distribution as a result, which has initial bursts on short rain events and centralized bursts on long rain events. The rainfall–runoff mechanisms vary depending on the burst differences. Another factor is that, given the study area's capability for drainage with the application of LID, the probability of rain is relatively low compared to the study's probability of rain. Many researchers agree that the most likely duration, frequency, and high rainfall patterns are sensitivities to the application of LID (Jemberie & Melesse 2021). That is what causes NPV scenarios I2, I3, and I4 to be negatively valued because the probability of rain is very low, that is, the depth of rain of 25 mm is 25 days, 30 mm is 15 days and 35 mm is 5 days. So, it can be said that although the depth of rainfall is small, the frequency of the rain is high, it provides greater benefits than the depth of high rain, but the frequency of occurrence is small.

Based on hydrological performance, scenario I shows a higher runoff reduction performance than scenario WS. The depth of rainfall affects the potential of LID in reducing the runoff, where the higher rainfall depth is, the lower runoff rate decreased. Based on the benefits aspects, scenario I provides greater benefits with more cost than the WS scenario. In terms of LCC, the payback period of the WS scenario is faster than scenario I, where the cost-effectiveness of the WS scenario is greater than the scenario I with a B/C above 1, especially in rainfall of 20 mm. This is due to the rainfall probability is more frequent than other rainfall depth. Thus, the most optimal LID design based on hydrological performance and benefit–cost analysis is the WS scenario with a rainfall depth of 20 mm. The results of the sensitivity analysis show that B/C is very vulnerable to changes in cost and discount rate. Many benefits can be received by applying LID in conserving water resources at the University of Jember, especially by using RBs. This research can be continued by assessing the integrated use of a combination of RBs, VSs, and bioretention that have not been considered here, as well as the use of other LID designs such as permeable pavements, green roofs, and others.

The authors want to share gratitude toward Jember University for financing this research.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.